torch tensorrt online with Winfy

We have hosted the application torch tensorrt in order to run this application in our online workstations with Wine or directly.

Quick description about torch tensorrt:

Torch-TensorRT is a compiler for PyTorch/TorchScript, targeting NVIDIA GPUs via NVIDIA�s TensorRT Deep Learning Optimizer and Runtime. Unlike PyTorch�s Just-In-Time (JIT) compiler, Torch-TensorRT is an Ahead-of-Time (AOT) compiler, meaning that before you deploy your TorchScript code, you go through an explicit compile step to convert a standard TorchScript program into a module targeting a TensorRT engine. Torch-TensorRT operates as a PyTorch extension and compiles modules that integrate into the JIT runtime seamlessly. After compilation using the optimized graph should feel no different than running a TorchScript module. You also have access to TensorRT�s suite of configurations at compile time, so you are able to specify operating precision (FP32/FP16/INT8) and other settings for your module.

Features:

• Build a docker container for Torch-TensorRT
• NVIDIA NGC Container
• Requires Libtorch 1.12.0 (built with CUDA 11.3)
• Build using cuDNN & TensorRT tarball distributions
• Test using Python backend
• You have access to TensorRT's suite of configurations at compile time

Programming Language: C++.
Categories: